What the work actually involves

You are given a workflow instruction — build a landing page report segmented by channel, diagnose a drop in a conversion event, set up an exploration comparing returning versus new users — and you complete it inside a live Google Analytics property while recording your screen. The recording is half the deliverable. The other half is written: a realistic prompt that a user might plausibly type to ask for this task, a step-by-step account of what you clicked and why, and a rubric that lets a reviewer grade an AI's attempt at the same task as correct, partially correct, or wrong. Some days are production; others are review, where you check another contributor's output against project guidelines and flag steps that were skipped, mislabeled, or done in a way GA4 doesn't actually support.

What the screen is looking for

micro1 runs an AI-led interview, and the probes go after depth that only shows up with hands-on GA4 use. Expect follow-ups on the difference between Universal Analytics and GA4's event-based model, what a key event is and how it's configured, how Explorations differ from standard reports, where data thresholding and sampling bite, and why two reports on the same property can disagree. Prior human-data work — annotation, RLHF, model output evaluation — is a strong signal but not required; what matters more is whether you can explain a workflow clearly enough that someone else could reproduce it, and whether you can write a rubric that distinguishes a near-miss from a real failure.

Logistics

  • Contractor engagement, fully remote, asynchronous task queues rather than fixed shifts.
  • You must supply your own Google Analytics access — a property with real or demo traffic you can navigate independently. Read-only access to a client account you're not authorized to record is a problem; raise it early.
  • Screen recording is mandatory, so you need a stable machine, decent upload bandwidth, and a workspace where nothing confidential appears on screen.
  • Pay in this category has been observed between $30 and $100/hr depending on depth of GA4 experience, evaluation background, and project tier. Nothing about that band is guaranteed.
  • Volume fluctuates with customer demand; treat it as supplementary rather than a guaranteed weekly load.